解决目标检测中特征融合错位问题，提升实时检测精度
Rethinking Features-Fused-Pyramid-Neck for Object Detection
- 提出独立层级金字塔结构，避免多尺度特征强制对齐
- 采用软最近邻插值与自适应选择机制，保留关键纹理细节
- 适合追求高精度实时检测的工业应用，尤其轻量化部署
多头检测器普遍采用特征融合的金字塔颈部进行多尺度检测，但在不同层级特征强行点对点融合时易出现特征错位。为此，本文设计了独立层级金字塔（IHP）架构，评估无融合金字塔颈部的有效性；引入带权重衰减因子的软最近邻插值（SNI），缓解跨层级融合带来的干扰，同时保留关键纹理信息；进一步提出扩展空间窗口中的特征自适应选择方法（ESD），增强轻量级卷积技术（GSConvE）。上述改进构成二级特征对齐方案（SA），在Pascal VOC和MS COCO上达到领先性能。代码将发布于https://github.com/AlanLi1997/rethinking-fpn。论文已获ECCV2024接收并由Springer Nature出版。
原文摘要 · Abstract (English)
Multi-head detectors typically employ a features-fused-pyramid-neck for multi-scale detection and are widely adopted in the industry. However, this approach faces feature misalignment when representations from different hierarchical levels of the feature pyramid are forcibly fused point-to-point. To address this issue, we designed an independent hierarchy pyramid (IHP) architecture to evaluate the effectiveness of the features-unfused-pyramid-neck for multi-head detectors. Subsequently, we introduced soft nearest neighbor interpolation (SNI) with a weight downscaling factor to mitigate the impact of feature fusion at different hierarchies while preserving key textures. Furthermore, we present a features adaptive selection method for down sampling in extended spatial windows (ESD) to retain spatial features and enhance lightweight convolutional techniques (GSConvE). These advancements culminate in our secondary features alignment solution (SA) for real-time detection, achieving state-of-the-art results on Pascal VOC and MS COCO. Code will be released at https://github.com/AlanLi1997/rethinking-fpn. This paper has been accepted by ECCV2024 and published on Springer Nature.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。